health-trend-analyzer
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill aggregates health data from a wide variety of JSON files across multiple subdirectories to identify patterns and generate insights. This ingestion of external data without explicit sanitization creates a vulnerability surface.
- Ingestion points: Data is read from
data/profile.json,data/symptoms/**/*.json,data/mood/**/*.json,data/diet/**/*.json,data/medication-logs/**/*.json,data/cycle-tracker.json,data/pregnancy-tracker.json,data/menopause-tracker.json,data/allergies.json, anddata/radiation-records.json. - Boundary markers: The instructions do not specify any delimiters or safety markers to differentiate between data content and instructions during the analysis phase.
- Capability inventory: The skill utilizes the
Read,Grep,Glob, andWritetools to process data and generate HTML reports. - Sanitization: There is no evidence of input validation or content sanitization to prevent malicious instructions embedded in health records from influencing the agent's behavior during report generation.
Audit Metadata